The state of Quicksilver

[fogboy]

To Nicholas/Alcor or someone else involved in Quicksilver development:

Quicksilver is far-and-away the most useful, most productivity-
enhancing program available for the Mac. Nothing even comes close, and
yet at the moment it is fraught with several problems that it's devout
users have posted the details of in this group.

Can you please reply in this thread to clarify the current state of
Quicksilver development and its prospects?

Specifically, many of us are under the assumption that, since the
google tech talk and other mentions of it, Quicksilver is no longer in
active development, meaning that NJ/alcor is not writing new/changed
code to deal with problems that have surfaced since 10.5 (but there's
been no official word).

Here's the question: is anyone else out there working on Quicksilver,
now that it's open source?

If the answer is yes, can they please reply here with details on the
prospects for stabilizing it for 10.5 and/or future versions?

If the answer is no, is it because there's not a developed market and
therefore none of the folks who could do the coding can spare the
time, knowing that the hours spent would not be compensated?

If that's the case, perhaps we could collect replies in this thread
from those of us who'd be willing to commit to paying for a newly
updated and 10.5-stable version of Quicksilver.

So please reply here if you a) can clarify exactly what is happening,
by whom, in Quicksilver development; or b) if you know that you would
pay for a new, 10.5-stable version.

I, for one, would gladly pay more than any donationware/shareware i've
paid for to date (which have ranged from as little as $8 to as much as
$40), since Quicksilver is so much more useful and relied upon than
any of them.

Thanks to Nicholas and all the folks replying to threads with
solutions and expertise thus far, know that your efforts are so
appreciated.

[brad....@gmail.com]

If someone would commit to fixing the current issues and maintaining
the software, I'd throw down 40 clams for a retail product in a
heartbeat. QS is just that useful.

I don't have time to be hacking on it myself, but I'd also be willing
to help out in some administrative/operational capacity - scrum-
mastering or issues prioritization or hosting stuff or whatever.

I'll watch this thread.

- Brad Heintz

[Jono]

Ankur Kothari has been working on this build of Quicksilver cleaning
up the code in the app.

http://lipidity.com/apple/cleaning-up-quicksilver/
http://lipidity.com/apple/quicksilver-b5x-download/
http://lipidity.com/software/quicksilver/

At the moment he's not doing any more work on Quicksilver as he's busy
studying. Hopefully he'll be able to more when he gets some free time.

At the time NJ released the source code for app he didn't release the
code for the plug-ins.
(I'm only guessing at this but) I think that if the source was
released for the plug-ins & someone worked on those maybe that would
help with stability.

As I understand it NJ has stopped all work on the B5X branch of
Quicksilver & has no plans on resuming (Please correct me if I'm
wrong).

He's working on (or has been working on) a new version of Quicksilver.
But I'm a little unclear if he's continuing with it, stopped working
on it, or has any plans to continue with it?

Maybe NJ could update us on the state of that at the moment, & his
plans for the future?


With regards to donations/funding Quicksilver I (along with plenty of
other people saying so in the past) would be more than willing to
donate or pay for Quicksilver if it meant it was being actively
developed, stabilised, existing features improved, or more features
added.

[Howard Melman]

FYI, I plan on diving into the code as soon as I get a new machine
with Leopard. I'm just waiting for macbook pro's to get updated. I
have a lot of programming experience but I'm pretty new to cocoa and
obj-c and QS doesn't seem to be the easiest codebase as an
introduction, but I'll give it a shot.

Howard

[Jono]

Great news Howard. Where do we send the money? :)

[Calvin]

its an interesting topic - open-source development post the original developers.  i've had some experience in this with releasing products i've written as open-source hoping someone would pick up development and drive it forward only to be dissapointed in that the product stagnates.  i can only put it down to that i think 95% if not all of the users are not developers.  for those that are developers its not always that simple to just dive into the code.  the original developer(s) are intimately familiar with the code base and could fix bugs and implement new features much faster.  if you do get people taking up the development then there is no singular force taking up the reigns and multiple branches happen with everyone with their own ideas.  the fact that quicksilver is distributed through svn will help that somewhat.

i don't think future releases built on this code base should be sold.  it would be unfair to NJ/alcor considering the immense foundation they have built.  money is not always motivation to a good product.  possibly thats what NJ/alcor has in mind for this 'new' quicksilver which is maybe being built to be a commercial product from the start?

i personally find quicksilver indispensable in daily life and also suffer frequent crashes.  i happen to be a developer but obj-c/cocoa is very new to me.  i would love to be able to contribute but as with most developers reading this i think time is the major factor.

calvin.

[Jono]

On Jan 29, 6:38 am, Calvin <c35t...@gmail.com> wrote:
> i can only
put it down to that i think 95% if not all of the users are not
developers.

Yea, I think that's the main problem. Most of us would dive in & help
but aren't developers & know how to code.

[fogboy]

thanks calvin this was well put, and really helpful to help us better
understand what's happening. it's kind of what i thought might be
happening, but i wasn't sure. i was hoping there was someone who had
been closer to the development process (someone who knew the code
nearly as well as NJ) who might be interested, but for whom the only
impediment was time/money.

NJ, any chance you could give us an idea of what the plan is for this
new iteration of QS?

thanks

On Jan 28, 10:38 pm, Calvin <c35t...@gmail.com> wrote:
> its an interesting topic - open-source development post the original
> developers. i've had some experience in this with releasing products i've
> written as open-source hoping someone would pick up development and drive it
> forward only to be dissapointed in that the product stagnates. i can only
> put it down to that i think 95% if not all of the users are not developers.
> for those that are developers its not always that simple to just dive into
> the code. the original developer(s) are intimately familiar with the code
> base and could fix bugs and implement new features much faster. if you do
> get people taking up the development then there is no singular force taking
> up the reigns and multiple branches happen with everyone with their own
> ideas. the fact that quicksilver is distributed through svn will help that
> somewhat.
>
> i don't think future releases built on this code base should be sold. it
> would be unfair to NJ/alcor considering the immense foundation they have
> built. money is not always motivation to a good product. possibly thats
> what NJ/alcor has in mind for this 'new' quicksilver which is maybe being
> built to be a commercial product from the start?
>
> i personally find quicksilver indispensable in daily life and also suffer
> frequent crashes. i happen to be a developer but obj-c/cocoa is very new to
> me. i would love to be able to contribute but as with most developers
> reading this i think time is the major factor.
>
> calvin.
>

[AIcor]

So here goes:

I use QS all day long and spend most, if not all, of my free time
working on Quicksilver (related) things--just not the branch.

Compensation is not a driver for me, and while I would like to
encourage other developers to contribute, I don't want money to be the
motivation. I would like any proceeds to go to charity and I may set
up a foundation to facilitate this.

I really want the branch to die. It has problems beyond simple bug
fixes.

The plug-in source has not been released because many of them have
already been converted to work with a new infrastructure. It is
important for me to drive development towards that.

Timelines are not important to me.

Any version of Quicksilver I make (including stuff I'm working on now)
will be free. I'm not developing a commercial version.

Consider who I chose to be my employer.

)|(

[Calvin]

from your comments below on the branch and 'new' versions you're
working on it doesn't seem as if you're abandoning the project, which
is probably what most people are thinking, but are rather working
towards a better solution that can be sustainable. with that in mind
and the fact that there is a future for this would it not be wise to
familiarise people with the new code base now and where its headed to,
and work towards a product that is driven forward primarily by
yourself but also able to be collectively developed. it would take
alot of the 'burden' off you for being the only point of development
for the product. only downside is the administration which can
sometimes be a larger task than the development.

whilst you wish the branch to disappear i believe it needs to get to a
point of stability, no new features should be implemented. reasoning
is that there is a large user-base that is fully behind this and their
support/interest in quicksilver needs to be maintained. user support
can quickly be dashed by frequent problems not being addressed. this
is not necessarily your task but can be done by developers wanting to
lend a hand. we do however need your input as to where in your
opinion the bugs lie. if they're in the released QS base then we can
all get that so its sorted but if it lies within unreleased plug-ins
then those need to be opened too. is there a specific reason besides
the new plugin architecture that you won't release the old plugins?

i greatly appreciate that the current and future quicksilver will be
free.

calvin.

[Rosso]

Alocr:
What do U think about whether Apple will further develop
spotlight into something functionally similar to Quicksilver?

[Jono]

Thanks for the Update Alcor.

I guess the problem most people have is if the new version of
Quicksilver has no ETA then they want (need?) a more stable version of
the current version to tie them over until the new version surfaces.
So it's good that other people (like Ankur) have got involved to try
to stabilise things (as it looks like it will be quite a while until
we get the new version), freeing you up to work on the new version.

Are you able to tell us how the new version will differ from the
current version?
(If you don't want to let us in on new features, how about a little
explanation on how you're developing it? I.E. Are you trying to build
it from the ground up so that it's more stable, the code is cleaner,
easier to add new features in the future etc.)

[AIcor]

Rosso: I don't know what Apple will do. I think it is unlikely, but
possible.

Calvin: Encouraging progress is the main reason I'm not releasing the
plug-ins. I'm happy to consider releasing each plug-in on a case by
case basis. (File tags and address book plugins were open sourced
because they were particularly unstable)

Jono: Mainly I am reconsidering all the original architecture
decisions. I want it to be more flexible and not as tied to a single
application.

[fogboy]

first, just a quick thank you to Alcor and others, this is exactly the
kind of illuminating exchange i was hoping for in this thread.

second, while i (and others i know) would indeed really like to have a
more stable and functional QS while we are waiting for the new
version, simply knowing that a stable, redesigned qs is coming at some
point in the not-too-distant future makes the restarts and loss of
certain functions that much easier to bear. (and hopefully, NJ, you
don't read that as trying to compel you to nail anything down as to
what "not-too-distant" menas, just saying that it helps me and the
folks i work with and train to know that there is indeed light at the
end of the tunnel.)

thanks.

[nontoppo]

On Jan 30, 12:30 pm, Jono <iconaho...@gmail.com> wrote:
> Are you able to tell us how the new version will differ from the
> current version?

The Google SVN trunk has a readme that goes into considerable detail
as to the new infrastructure. By the sounds of it, plugins are using a
much more robust interface, triggers will be moved out of QS into a
global app, and all the little accesory functions are being rolled up
into a larger framework.

I wish more developers were able to hammer out all the Leopard kinks
in the current build until QS II emerges from its shell. Ankur made a
small start, but the core bugs are still around (great news that file
tag is open to stop crashing, thanks alcor!). Perhaps a bounty system
for bug fixes would work to "encourage" coders to dive in with
stopgaps?

[seyDoggy]

I know it's been said already but Ankur over at Lipidity.com has
hammered out the next unstable branch with some major clean-up and
some bug fixes. He's released from his site a B55 release. It's a bit
tricky to get it fly from the start (there are a few terminal commands
to get all the privileges necessary). In the end though it feels much
faster and lighter on it's feet.

[fogboy]

Thanks a ton seyDoggy! -- i posted detailed steps to take here in the
google discussions:

http://groups.google.com/group/blacktree-quicksilver/browse_thread/thread/d41906df20c7361b/476c72d307d72eb4#476c72d307d72eb4

[robertso...@gmail.com]

Well, I hope that Quicksilver lives on.
I suppose it wouldn't be too bad if Apple took over the job. Anybody
feel like sending them feedback? :P

Good luck Alcor!

On Feb 1, 4:03 pm, fogboy <fogboy...@gmail.com> wrote:
> Thanks a ton seyDoggy! -- i posted detailed steps to take here in the
> google discussions:
>

> http://groups.google.com/group/blacktree-quicksilver/browse_thread/th...

[Kevin]

Alcor, would you consider cleaning up or releasing fixed versions of
the file compression and user interface plugins for Leopard? Being
unable to use the comma trick to create a zip archive is a big snag in
my workflow (though I believe it's been that way for a while), as is
the brokenness of the "Show Menu Items" action. I don't know if the
plugins require significant changes to be fixed, but they were two of
the most beloved for a lot of people, and would help tide everyone
over until your re-envisioning of Quicksilver is complete.

On Feb 7, 2:20 am, "robertson.patr...@gmail.com"

[Bruno Abreu]

Does anybody notice a stability improvement in Quicksilver under
10.5.2, or is it just a misperception of mine?

[seyDoggy]

It seems more stable, but less proxy object work. :(

[Tim Jeffries]

I'm finding it more stable and maybe not as sluggish either. That said
I've probably jinxed myself by posting that.

[David Chartier]

On Jan 29, 10:12 am, AIcor <j...@gmail.com> wrote:
> Compensation is not a driver for me, and while I would like to
> encourage other developers to contribute, I don't want money to be the
> motivation. I would like any proceeds to go to charity and I may set
> up a foundation to facilitate this.
>

Thanks for the responses on this list Alcor. They certainly do help.

However, It sounds like you may be misunderstanding the offers from
enthusiastic users to offer compensation for Quicksilver as some kind
of "bribe." I think it's safe to say that when most of us make an
offer like that (including myself), it's under the assumption that you
have some kind of a day job at which you can't program for
Quicksilver. The idea is that if you could start making enough money
through donations or by charging for a commercial license (even when
offering QS as an open source app, if you prefer), you'd be able to
get rid of your non-developer job and write Quicksilver part time or
even full time.

I'm in the same boat as many other users here: I depend on Quicksilver
for so much of my productivity, as it's arguably one of the greatest
applications ever invented. But the bugs and instability are getting
painful to deal with, and I genuinely be delighted to make a
significant donation or--preferably--purchase a commercial license if I
knew you or another group of developers were supporting Quicksilver
full time.

Quicksilver is an incredibly significant, powerful, and influential
application, and we would love to see its developers be able to
develop it to its full, polished potential. That's all the
compensation offers are about.

[seyDoggy]

I think David has said it well. I don't think Alcor is about to change
his mind but I wish he would consider selling the code base or opening
it fully to allow others to take this to the next level. The supported
alternatives (like launchbar) just don't cut it. While they feel like
an app, QS feels like your operating system.

My current setup on B5X from http://lipidity.com/apple/quicksilver-b5x-download/
is the most stable and most responsive I've had yet but the lack of
proxy objects in Leopard is killing me. To bring all of the features
back, I too wish I could have a paid-for development of QS. It is the
app I rely on most. Nearly every keystroke depends on QS. It is an
extension of my brain. Don't let my brain wither away and die.

On Mar 15, 10:55 pm, David Chartier <dcha...@gmail.com> wrote:
> On Jan 29, 10:12 am, AIcor <j...@gmail.com> wrote:
>

> > Compensation is not a driver for me, and while I would like to
> > encourage other developers to contribute, I don't want money to be the
> > motivation. I would like any proceeds to go to charity and I may set
> > up a foundation to facilitate this.
>

> Thanks a lot for replying here Alcor. Your interaction really helps to
> understand QS's progress.
>
> However, I hope you aren't taking the offers for compensation the
> wrong way, or as some kind of "bribe."
>
> I think it's fair to say that the people making these offers,
> including myself, assume that you have some kind of a job (apart from
> your studies) in which you can't develop Quicksilver. The idea behind
> offering compensation is that, if we could give you enough, help get
> the word out, and get Quicksilver to start bringing in a steady stream
> of money, you (or perhaps another developer or team) could afford to
> cut back on the day job (or quit it), and work on Quicksilver in a
> more steady fashion.

>
> Quicksilver is an incredibly significant, powerful, and influential

> application that a great many of us depend on. Our offers for
> compensation really are just attempts to give Quicksilver's developers
> the time to polish it into the serious application it deserves to be.

[Hagure]

Would you stop working at google to become a shareware developer?

Hehe... a mac user can dream...

seyDoggy: Most proxy items work for me in Leopard... have you done the
QS (or even Leopard) reinstall dance?

[Luhmann]

Having read the FAQ and this thread, I *still* feel confused about the
state of Quicksilver. If I understand correctly, this is the current
situation:

1. The developer, Alcor decided that the original code base was too
messed up, so announced he would no longer be developing that
particular version of QS.

2. This was widely misinterpreted as meaning that QS is dead - but
that is incorrect. In fact, Alcor is simply working on a new version
of QS which will be even better than the old one.

3. Meanwhile, a lot of Leopard users are frustrated by the performance
of the current version of QS, and Ankur Kothari has taken it upon
himself to make improvements to the current code base. Those can be
found here:

http://lipidity.com/software/quicksilver/

However, while some people have been able to make it work, it is not
recommended for the average user, and breaks some features of QS.

4. So the end result is that current QS users have a choice between
the official unstable version and an unofficial unstable version, but
at some unspecified distant time in the future Alcor will release a
new and improved version which he is working on when he isn't at his
day job with Google. Money will not speed this process along, nor will
complaining on the forums. So ... patience.

Is this correct?

[Howard Melman]

That seems about right. I'll add that QS has always been in beta and
some things have, as near as I can tell, never worked right. Saying
QS is "unstable" is probably a bit strong. Some plugins aren't
working that used to work (I tried to start a thread to track them but
got very little feedback) and some (like iTunes) work better. If a
plugin you use regularly is broken, then it's unstable for you, but
given that everyone uses different plugins, it might be just fine for
someone else. When things have broken badly (like the site going
down) Alcor did fix things reasonably quickly, so unsupported is too
strong a term too.

Howard

[seyDoggy]

Just to chime in on #2:
I don't know that Alcor is work on a new version per say. I think he
is further developing certain components or aspects from QS into
something entirely different all together. But that was just my take
on it.

[Luhmann]

Yeah. I still use it every day, so it works OK for me - but
performance seems less reliable since Leopard ...

[pendolino]

I've been using it quite well with leopard but note that i have B53
(3814) running and will note upgrade since all is well and i dont want
to break anything.

[Luhmann]

I was able to greatly improve the performance under Leopard by
replacing the Bezel interface with this one:

http://www.mygnu.com/julius/proj_silverflow.html

Although beta, it is much snappier in Leopard than the built-in bezel
theme. Looks nice, and seems stable.

[oli]

I was trying to find a h4xr.org plugin for quicksilver since
konfabulator / dashboard widget hasnt been my thing since the
discovery of QS.

And now i find this...I've been wondering why new version of
quicksilver hasnt showed up on macupdate for a long time, albeit it
took me like a year to figure out something's wrong.

I'm deeply saddened by this. I've been using QS since Panther. Hope
someone picks up the project or Alcor comes up with a new rewritten
version.